tcp: tcp challenge ack fixes
Subsystems (4)
cover tcp: tcp challenge ack fixes
From: Eric Dumazet <[email protected]>
syzbot found a typical data-race addressed in the first patch.
While we are at it, second patch makes the global rate limit per net-ns and disabled by default.
Eric Dumazet (2):
tcp: annotate data-race around challenge_timestamp
tcp: make global challenge ack rate limitation per net-ns and default
disabled
Documentation/networking/ip-sysctl.rst | 5 ++++- include/net/netns/ipv4.h | 2 ++ net/ipv4/tcp_input.c | 21 +++++++++++---------- net/ipv4/tcp_ipv4.c | 6 ++++-- 4 files changed, 21 insertions(+), 13 deletions(-)
Signature · 1 line
2.37.2.672.g94769d06f0-goog
PATCH 1/2 tcp: annotate data-race around challenge_timestamp
From: Eric Dumazet <[email protected]>
challenge_timestamp can be read an written by concurrent threads.
This was expected, but we need to annotate the race to avoid potential issues.
Following patch moves challenge_timestamp and challenge_count to per-netns storage to provide better isolation.
--- net/ipv4/tcp_input.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
net/ipv4/tcp_input.c +2 -2
@@ -3629,11 +3629,11 @@ static void tcp_send_challenge_ack(struct sock *sk)/* Then check host-wide RFC 5961 rate limit. */ now = jiffies /HZ ; -if (now != challenge_timestamp) { +if (now !=READ_ONCE (challenge_timestamp)) {u32 ack_limit =READ_ONCE (net->ipv4.sysctl_tcp_challenge_ack_limit);u32 half = (ack_limit +1 ) >>1 ; - challenge_timestamp = now; +WRITE_ONCE (challenge_timestamp, now);WRITE_ONCE (challenge_count, half +prandom_u32_max (ack_limit)); } count =READ_ONCE (challenge_count);
Signature · 1 line
2.37.2.672.g94769d06f0-goog
PATCH 2/2 tcp: make global challenge ack rate limitation per net-ns and default disabled
In reply to Eric DumazetFrom: Eric Dumazet <[email protected]>
Because per host rate limiting has been proven problematic (side channel attacks can be based on it), per host rate limiting of challenge acks ideally should be per netns and turned off by default.
This is a long due followup of following commits:
083ae308280d ("tcp: enable per-socket rate limiting of all 'challenge acks'") f2b2c582e824 ("tcp: mitigate ACK loops for connections as tcp_sock") 75ff39ccc1bd ("tcp: make challenge acks less predictable")
--- Documentation/networking/ip-sysctl.rst | 5 ++++- include/net/netns/ipv4.h | 2 ++ net/ipv4/tcp_input.c | 21 +++++++++++---------- net/ipv4/tcp_ipv4.c | 6 ++++-- 4 files changed, 21 insertions(+), 13 deletions(-)
Documentation/networking/ip-sysctl.rst +4 -1
@@ -1035,7 +1035,10 @@ tcp_limit_output_bytes - INTEGER tcp_challenge_ack_limit - INTEGER Limits number of Challenge ACK sent per second, as recommended in RFC 5961 (Improving TCP's Robustness to Blind In-Window Attacks) - Default: 1000 + Note that this per netns rate limit can allow some side channel + attacks and probably should not be enabled. + TCP stack implements per TCP socket limits anyway. + Default: INT_MAX (unlimited) UDP variables =============
include/net/netns/ipv4.h +2 -0
@@ -179,6 +179,8 @@ struct netns_ipv4 {unsigned int sysctl_tcp_fastopen_blackhole_timeout;atomic_t tfo_active_disable_times;unsigned long tfo_active_disable_stamp; +u32 tcp_challenge_timestamp; +u32 tcp_challenge_count;int sysctl_udp_wmem_min;int sysctl_udp_rmem_min;
net/ipv4/tcp_input.c +11 -10
@@ -3614,12 +3614,9 @@ bool tcp_oow_rate_limited(struct net *net, const struct sk_buff *skb,/* RFC 5961 7 [ACK Throttling] */ static void tcp_send_challenge_ack (struct sock *sk) { -/* unprotected vars, we dont care of overwrites */ -static u32 challenge_timestamp; -static unsigned int challenge_count;struct tcp_sock *tp =tcp_sk (sk);struct net *net =sock_net (sk); -u32 count, now; +u32 count, now, ack_limit;/* First check our per-socket dupack rate limit. */ if (__tcp_oow_rate_limited(net,
@@ -3627,18 +3624,22 @@ static void tcp_send_challenge_ack(struct sock *sk) &tp->last_oow_ack_time))return ; + ack_limit =READ_ONCE (net->ipv4.sysctl_tcp_challenge_ack_limit); +if (ack_limit ==INT_MAX ) + goto send_ack; +/* Then check host-wide RFC 5961 rate limit. */ now = jiffies /HZ ; -if (now !=READ_ONCE (challenge_timestamp)) { -u32 ack_limit =READ_ONCE (net->ipv4.sysctl_tcp_challenge_ack_limit); +if (now !=READ_ONCE (net->ipv4.tcp_challenge_timestamp)) {u32 half = (ack_limit +1 ) >>1 ; -WRITE_ONCE (challenge_timestamp, now); -WRITE_ONCE (challenge_count, half +prandom_u32_max (ack_limit)); +WRITE_ONCE (net->ipv4.tcp_challenge_timestamp, now); +WRITE_ONCE (net->ipv4.tcp_challenge_count, half +prandom_u32_max (ack_limit)); } - count =READ_ONCE (challenge_count); + count =READ_ONCE (net->ipv4.tcp_challenge_count);if (count >0 ) { -WRITE_ONCE (challenge_count, count -1 ); +WRITE_ONCE (net->ipv4.tcp_challenge_count, count -1 ); +send_ack:NET_INC_STATS (net,LINUX_MIB_TCPCHALLENGEACK );tcp_send_ack (sk); }
net/ipv4/tcp_ipv4.c +4 -2
@@ -3139,8 +3139,10 @@ static int __net_init tcp_sk_init(struct net *net) net->ipv4.sysctl_tcp_tso_win_divisor =3 ;/* Default TSQ limit of 16 TSO segments */ net->ipv4.sysctl_tcp_limit_output_bytes =16 *65536 ; -/* rfc5961 challenge ack rate limiting */ - net->ipv4.sysctl_tcp_challenge_ack_limit =1000 ; + +/* rfc5961 challenge ack rate limiting, per net-ns, disabled by default. */ + net->ipv4.sysctl_tcp_challenge_ack_limit =INT_MAX ; + net->ipv4.sysctl_tcp_min_tso_segs =2 ; net->ipv4.sysctl_tcp_tso_rtt_log =9 ;/* 2^9 = 512 usec */ net->ipv4.sysctl_tcp_min_rtt_wlen =300 ;
Signature · 1 line
2.37.2.672.g94769d06f0-goog
Re: [PATCH net 1/2] tcp: annotate data-race around challenge_timestamp
In reply to Eric DumazetQuoted 7 lines · Eric Dumazet
From: Eric Dumazet <[email protected]>
challenge_timestampcan be read an written by concurrent threads.This was expected, but we need to annotate the race to avoid potential issues.
Quoted 7 lines
Following patch moves
challenge_timestampandchallenge_countto per-netns storage to provide better isolation.Fixes 354e4aa391ed ("tcp: RFC 5961 5.2 Blind Data Injection Attack Mitigation")Reported-by syzbot <[email protected]>Signed-off-by Eric Dumazet <[email protected]>---
Thanks, Eric!
neal
Re: [PATCH net 2/2] tcp: make global challenge ack rate limitation per net-ns and default disabled
In reply to Eric DumazetQuoted 9 lines · Eric Dumazet
From: Eric Dumazet <[email protected]>
Because per host rate limiting has been proven problematic (side channel attacks can be based on it), per host rate limiting of challenge acks ideally should be per netns and turned off by default.
This is a long due followup of following commits:
Quoted 8 lines
083ae308280d("tcp: enable per-socket rate limiting of all 'challenge acks'")f2b2c582e824("tcp: mitigate ACK loops for connections astcp_sock")75ff39ccc1bd("tcp: make challenge acks less predictable")Signed-off-by Eric Dumazet <[email protected]>Cc Jason Baron <[email protected]>Cc Neal Cardwell <[email protected]>---
Looks great. Thanks, Eric!
neal
This series was applied to netdev/net.git (master) by Jakub Kicinski <[email protected]>:
Hello:
This series was applied to netdev/net.git (master) by Jakub Kicinski <[email protected]>:
Quoted 8 lines · On Tue, 30 Aug 2022 11:56:54 -0700 you wrote:
From: Eric Dumazet <[email protected]>
syzbot found a typical data-race addressed in the first patch.
While we are at it, second patch makes the global rate limit per net-ns and disabled by default.
[...]
Here is the summary with links:
- [net,1/2] tcp: annotate data-race around challenge_timestamp
https://git.kernel.org/netdev/net/c/8c70521238b7 - [net,2/2] tcp: make global challenge ack rate limitation per net-ns and default disabled https://git.kernel.org/netdev/net/c/79e3602caa6f
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html